This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Evaluate Effects of Personalized Patient Counselling for Enbrel® Therapy in Adults With Rheumatoid Arthritis (NCT02346877), a non-randomized phase 4 study. The primary outcome is Persistence of Study Drug Measured at 52 Weeks.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02346877 |
| Phase | Phase 4 |
| Condition | Persistence to Enbrel |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 19 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Amgen |
| Results posted | Yes |
2. Clinical Question
Population
Adults with rheumatoid arthritis receiving Enbrel (etanercept) in routine clinical practice. The registry record reports enrollment of 19 participants, but says no participants completed the study.
Intervention
Personalized patient counselling using an Information-Motivation-Strategy approach based on the Beliefs about Medicines Questionnaire. The registry record reports no participants in this cohort.
Comparator
A standard-of-care cohort intended to serve as a historical control. Participants in that cohort were to receive Enbrel in routine clinical practice.
Primary question
Would counselling improve persistence with Enbrel at 52 weeks compared with persistence estimated under standard care? The registry record says the efficacy endpoints were not evaluated because the study closed early and no participants completed it.
The question concerns persistence, not simply whether a participant started treatment or received a prescription. The planned endpoint used prescription supply, confirmation of injection in an electronic diary, and treatment gaps to classify persistence at 52 weeks. These elements describe the intended measure; they do not establish an outcome for this study.
The study was designed as a comparison between a standard-of-care cohort and a personalized-counselling cohort, but it was not randomized. The record describes the control cohort as preceding the counselling cohort, with counselling cohort initiation contingent on analysis of control participants. This structure makes the registry question different from a randomized comparison in which participants are assigned to groups under the same conditions. More importantly, the counselling cohort has no reported participants, and the planned efficacy comparison was not performed. The record therefore does not answer whether counselling changed persistence.
3. Trial Design
This was a phase 4, multicenter, non-randomized, unmasked, parallel study. Its official title describes a historical-control study of personalized patient counselling and persistence to Enbrel therapy in people with rheumatoid arthritis. The listed intervention was behavioral counselling; both cohorts were described as receiving Enbrel in routine clinical practice.
The registry description planned to enroll the first 100 participants in the standard-of-care control cohort. The personalized-counselling cohort was to begin after 75% of control participants had been analyzed and shown not to have high persistence. The record reports total enrollment of 19 and shows 19 participants at risk in the standard-of-care cohort and 0 in the counselling cohort for the posted adverse-event figures. It also says no participants completed the study. The actual conduct therefore did not yield the planned efficacy comparison.
The study began in 2015-04, had a primary completion date of 2015-10, and is listed as terminated. The reason for termination is not reported in the registry record. The record does not report a randomized allocation process, masking, or a concurrent control comparison; it lists allocation as non-randomized and masking as none.
A historical-control design can be useful for describing a clinical question when a concurrent randomized comparison is not used. However, differences between groups may reflect factors besides the intervention, including differences in timing or participant characteristics. That is a general limitation of this design, not evidence that any particular source of bias occurred. With no reported efficacy analyses and no participants in the counselling cohort, this record cannot estimate a comparative counselling effect.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Before assignment: Eligible participants were to be enrolled in the standard of care (control) cohort and complete the 52-week study period. Data from 75% of the group were to be analyzed before enrolling participants in the interventional cohort. As the study was prematurely closed due to slow enrollment, the interventional cohort was not enrolled.
Overall Study
| Stage | Standard of Care (Control) Cohort | Personalized Patient Counselling (Interventional) Cohort |
|---|---|---|
| STARTED | 19 | 0 |
| COMPLETED | 0 | 0 |
| NOT COMPLETED | 19 | 0 |
Reasons for not completing this period:
| Reason | Standard of Care (Control) Cohort | Personalized Patient Counselling (Interventional) Cohort |
|---|---|---|
| Adverse Event | 1 | 0 |
| Administrative Decision | 1 | 0 |
| Sponsor Decision | 12 | 0 |
| Missing | 5 | 0 |
Analysis population for the primary outcome, as defined in the registry: Efficacy endpoints were not evaluated because the study was prematurely closed and no participants completed the study.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | Standard of Care (Control) Cohort | Personalized Patient Counselling (Interventional) Cohort | Total |
|---|---|---|---|---|
| Participants analysed | 19 | 0 | 19 | |
| Age, Continuous | years | 57.1 (SD 10.1) | — | 57.1 (SD 10.1) |
| Sex: Female, Male: Female | Participants | 14 | — | 14 |
| Sex: Female, Male: Male | Participants | 5 | — | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Persistence of Study Drug Measured at 52 Weeks, measured over 52 weeks.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was persistence with study drug measured at 52 weeks. Under the registry definition, a participant would count as persistent if the last prescription, together with its weeks of supply, covered at least 52 weeks from enrollment and the electronic diary confirmed that the Enbrel from the prescription was injected. The definition also classified a participant as not persistent if supply ended before 52 weeks and the diary confirmed no injection, or if there was a treatment gap of more than 4 consecutive weeks for no medical reason during the study period.
Persistence is used to summarize whether treatment continues over a specified period, rather than only whether treatment was initiated. That is a general reason it can be relevant to a study of counselling intended to support ongoing therapy. Here, the planned endpoint combines prescription information with reported injection and a rule about treatment gaps, so it depends on those sources of information being available and interpretable.
The registry record says efficacy endpoints were not evaluated because the study was prematurely closed and no participants completed it. Thus, the 52-week definition is a protocol description, not an observed result. The record does not report a persistence estimate or a comparison between cohorts.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive point is that the registry record contains no efficacy result for the primary endpoint. Although persistence at 52 weeks was specified, the record says the study closed prematurely and no participants completed it. It lists 0 analyses. Accordingly, there is no reported persistence rate, treatment effect estimate, confidence interval, or hypothesis-test result to interpret.
The design also did not produce a like-for-like comparison. Allocation was non-randomized, and the planned control cohort was to precede the counselling cohort. The record reports 19 participants in the standard-of-care cohort for adverse-event reporting and none in the counselling cohort. This does not support a comparative estimate: there is no counselling cohort outcome with which to compare the control cohort, and the intended efficacy analysis was not conducted. A difference between the cohorts cannot be calculated from the posted efficacy results because those results are absent.
The sample-size material on the page is a planning illustration based on an observed effect, using stated assumptions about significance level, allocation, and a normal approximation. Such illustrations are not observed power and are not a substitute for trial results. The registry record does not report a primary endpoint effect from this study, so a reader should not treat a planning illustration as evidence that the trial detected or failed to detect a counselling effect. The illustration also does not establish what sample size the investigators originally planned or why.
The adverse-event figures are descriptive and concern a different question from persistence. The standard-of-care cohort has 0 serious events recorded among 19 at risk and 1 other adverse event among 19 at risk. The counselling cohort has 0 participants at risk in the posted figures, so its zero event counts do not indicate an absence of risk or establish safety. The death counts are not reported in the registry record. These figures do not make up for the absence of efficacy outcomes.
In general, a confidence interval describes the range of effect estimates compatible with the data under a statistical model. No confidence interval for persistence is reported. Likewise, a non-significant result, if one had been reported, would not by itself show that an intervention has no effect; the estimate and its precision would matter. This record has no efficacy estimate to assess for either magnitude or precision. The appropriate conclusion is limited: the planned clinical question is documented, but the posted record does not provide an answer.
9. Statistical Methods Explained
The registry describes the study as non-randomized, unmasked, and parallel, with a historical-control comparison. These features matter because the statistical meaning of an estimate depends on how the comparison groups were formed and what outcome data were collected. In a randomized study, assignment is intended to balance measured and unmeasured characteristics on average. In a non-randomized study, groups can differ for reasons other than the intervention. That is a general methodological concern; the record does not report which characteristics differed here, or whether the investigators adjusted for any of them.
What does the design permit?
The planned comparison was between standard care and personalized counselling, but the registry describes control participants as enrolled first and counselling cohort initiation as dependent on control-cohort analysis. A historical or sequential comparison may be affected by changes over time, differences in who enters each cohort, or other factors that coincide with the intervention. These are general possibilities, not established problems in this record. The registry lists no randomized allocation or masking. It also reports no counselling participants in the posted safety figures and no efficacy analyses, so there is no comparative endpoint estimate to interpret.
What would an effect estimate mean?
For a persistence endpoint, an analysis might summarize the proportion meeting the endpoint in each group or compare persistence between groups. The exact planned effect measure and statistical model are not reported in the registry record. Without an estimate, its uncertainty, and a description of the analysis population, readers cannot judge the size or precision of a treatment contrast. A confidence interval is generally used to show uncertainty around an estimate; it is not reported for the primary endpoint. A p-value is generally one measure of compatibility between observed data and a specified null model, but no primary endpoint p-value is reported either.
Why can over-dispersion matter?
General statistical explanation: for count outcomes, some simple models assume that variability follows a specified relationship to the mean. If observed variability is greater than that assumption allows, the data are over-dispersed. Ignoring over-dispersion can make standard errors too small and uncertainty appear narrower than it is. This concept does not establish that over-dispersion applies to this study. The registry record does not report an efficacy analysis, a count model, or a model assessment.
What do the planning illustrations show?
The page includes sample-size illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect. These are planning illustrations, not observed power and not evidence of an outcome. The record does not report the trial's own sample-size justification, and the primary efficacy endpoint was not evaluated. Readers should therefore distinguish the illustrative calculations from the study's design rationale and from any observed treatment effect.
What is the role of the endpoint definition?
The persistence definition combines prescription coverage, electronic-diary confirmation of injection, and a rule about gaps longer than 4 consecutive weeks for no medical reason. General measurement considerations include whether each source is complete, whether the timing is recorded consistently, and how ambiguous or missing information is handled. The registry record gives the endpoint rule but does not report how missing diary entries or other incomplete endpoint information would have been handled. Because no participants completed the study, the record says the efficacy endpoints were not evaluated.
Can the safety counts be analyzed comparatively?
The registry posts adverse-event counts by cohort, but the counselling cohort has 0 participants at risk. That means there is no observed counselling-cohort denominator from which to estimate its event frequency. The standard-of-care counts remain descriptive. General comparisons of safety require exposure information and sufficiently reported outcomes in the groups being compared; the posted counts do not provide a comparative safety analysis. No statistical analysis is listed in the record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists secondary outcomes for adherence to study drug up to 52 weeks and the Beliefs about Medicines Questionnaire. It also lists 3 outcome measures, but reports 0 analyses. The record does not provide results for these secondary measures or identify any statistical tests performed on them.
Multiplicity arises when several outcomes, time points, or comparisons are tested. In general, conducting many tests can increase the chance of at least one apparently notable result, so a statistical plan may specify how to interpret or adjust multiple tests. The registry record does not report a multiplicity strategy, a hierarchy of endpoints, or an analysis plan that would show how multiple outcomes were to be handled. It would be inappropriate to infer that an adjustment was or was not performed.
The record says no participants completed the study and that efficacy endpoints were not evaluated. It does not describe how missing outcome data would have been handled, how the analysis population was defined, or whether any interim analyses were planned. These details are not reported in the registry record. Because the efficacy analyses were not conducted, there are no posted results for which to assess the consequences of missing data or multiplicity.
These omissions limit statistical interpretation but should not be filled in by assumption. The record establishes that the primary and secondary outcome results are unavailable; it does not establish a particular reason beyond premature closure for their absence, nor does it describe the investigators' intended procedures for incomplete observations.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Standard of Care (Control) Cohort | 0 | 19 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Standard of Care (Control) Cohort | 1 of 19 |
| Personalized Patient Counselling (Interventional) Cohort | 0 of 0 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Standard of Care (Control) Cohort | Personalized Patient Counselling (Interventional) Cohort |
|---|---|---|---|
| Injection site reaction | Injury, poisoning and procedural complications | 1 of 19 | 0 of 0 |
The posted safety figures list the standard-of-care cohort with 19 participants at risk for serious adverse events and 0 serious events. For other adverse events, the same cohort has 19 participants at risk and 1 event. The personalized-counselling cohort has 0 participants at risk in both categories, with 0 events recorded in each. The death counts are not reported in the registry record, and no serious adverse-event terms are listed.
These counts are descriptive. In particular, zero events in a cohort with 0 participants at risk do not provide evidence about the event experience of people receiving counselling. The figures also do not establish a comparative safety profile: the record does not report an analysis comparing cohorts, and the design was non-randomized.
Serious adverse events and other adverse events are separate categories in the posted record. The single other event should not be read as a persistence result or as a treatment-effect estimate. The registry does not provide further detail about that event in the registry record, so its nature and relationship to treatment cannot be characterized.
12. What the Result Does, and Does Not, Mean
This record documents a study intended to assess whether personalized counselling could improve persistence with Enbrel therapy at 52 weeks relative to a standard-of-care historical control. It also documents a planned endpoint definition, the non-randomized cohort structure, and the status of the study. Those details explain the research question, but they do not constitute evidence of an effect.
The registry says the study was terminated and prematurely closed, with no participants completing it. It explicitly states that efficacy endpoints were not evaluated, and lists 0 analyses. Consequently, the posted record does not show whether persistence differed between cohorts, how large any difference might have been, or how precise an estimate would be. It cannot be interpreted as demonstrating benefit, demonstrating no benefit, or establishing equivalence.
The safety counts are similarly limited. They include 19 participants at risk in the standard-of-care cohort and none in the counselling cohort. The absence of counselling participants in those figures means they cannot answer what adverse events occurred in that cohort. The standard-of-care event counts do not fill the gap in the planned efficacy comparison.
General statistical interpretation distinguishes absence of evidence from evidence of absence. When no endpoint estimate is reported, there is no estimate or uncertainty interval with which to assess whether an effect is large, small, or compatible with no difference. The sample-size illustrations elsewhere on the page are planning calculations and are not trial findings or observed power. The registry also does not report the reason for termination, the trial's sample-size justification, or a detailed analysis plan. These limits should remain explicit when describing what this study contributes.
13. Why This Trial Matters Statistically
This trial is statistically informative as an example of the difference between a well-defined research question and an answer supported by analyzed outcomes. The registry gives a specific persistence definition and describes a counselling strategy based on questionnaire results. Yet it also states that the study closed early, no participants completed it, and efficacy endpoints were not evaluated. A detailed endpoint description cannot replace outcome data.
The record also illustrates why the structure of a comparison matters. The study was non-randomized and planned a standard-of-care cohort followed by a counselling cohort under a historical-control design. Without random assignment, differences between groups can reflect factors beyond counselling. That is a general interpretive issue, not a finding about this study's participants. In this case, the absence of a counselling cohort in the posted figures and the absence of efficacy analyses leave no comparative estimate to assess.
Finally, the record distinguishes descriptive safety counts from an efficacy comparison. A count in the standard-of-care cohort does not answer the primary persistence question, and a zero count for a cohort with no participants at risk is not a safety result for that cohort. These distinctions are useful when reading registry pages: planned questions, observed outcomes, and statistical analyses are separate parts of the evidence.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The main limitation is that the primary efficacy endpoint was not evaluated. The registry says no participants completed the study and lists 0 analyses, so there is no posted result for persistence at 52 weeks. The study is listed as terminated, but the reason for termination is not reported in the registry record.
The enrolled total was 19, while the registry description had planned a first 100 participants in the standard-of-care cohort. The posted adverse-event figures show 19 participants at risk in that cohort and 0 in the personalized-counselling cohort. There is therefore no counselling-cohort outcome in those figures, and the intended comparison cannot be assessed. The record does not explain how the observed enrollment came to differ from the described plan.
The non-randomized, historical-control design would also require caution even if comparative outcomes were available. General concerns include group differences and changes over time, but the record does not report evidence that any particular bias occurred. It does not report a statistical analysis plan, a sample-size justification, handling of missing data, an analysis population, or an approach to multiplicity. These details are not reported in the registry record.
The endpoint definition depends on prescription information and electronic-diary confirmation, but the record does not describe how incomplete or conflicting records would have been treated. The safety information is limited to cohort-level counts; the counselling cohort has no participants at risk, death counts are not reported, and no serious-event terms are listed. These limits prevent broad conclusions about either comparative safety or efficacy.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Evaluate Effects of Personalized Patient Counselling for Enbrel® Therapy in Adults With Rheumatoid Arthritis — NCT02346877.
- Registry record last updated 2017-03-27; status last verified 2017-03.
- Results first posted 2017-03-27.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
This phase 4 study examined personalized patient counselling alongside Enbrel therapy for adults with rheumatoid arthritis. It was listed as multicenter, non-randomized, unmasked, and parallel, with a historical standard-of-care control. The primary endpoint was persistence at 52 weeks, defined using prescription supply, electronic-diary confirmation, and treatment gaps.
The record reports enrollment of 19, a start date of 2015-04, a primary completion date of 2015-10, and terminated status. It says the study was prematurely closed and no participants completed it; efficacy endpoints were not evaluated, and 0 analyses are listed.
Safety figures list 19 participants at risk in the standard-of-care cohort and 0 in the counselling cohort. The record does not report a primary efficacy result, the reason for termination, the sample-size justification, or a detailed statistical analysis plan.